Agentforce Knowledge Action Degradation

Investigations have highlighted that the impact radius is broader than initially understood. We have updated the posting to reflect the additional instances experiencing impact, and customers in these instances will receive communications related to this incident going forward.

We have identified the trigger and are taking action to roll back the recent release. As of 14:41 UTC, the test was successful on ITA22S, and errors have cleared on this instance. At 14:51 UTC, a targeted rollback began for ITA6, ITA4S and GBR90S, which is estimated to take approximately one hour.

We’ll provide an update in 60 minutes or sooner if additional information becomes available.

We have reproduced the issue and have moved onto the process of rolling back the recent release on a test instance (ITA22S). We will validate whether this provides the desired results and brings the service back to a healthy state. If it is successful, we will roll back all impacted instances.

We’ll provide an update in 60 minutes or sooner if additional information becomes available.

We are bringing in additional expertise to reproduce the issue to determine whether a rollback or the recent release of a fix forward would be a viable option.

We’ll provide an update in 60 minutes or sooner if additional information becomes available.

We are attempting to reproduce the issue to determine whether or not the recent release is indeed the trigger of the issue and what would be the best course of action to mitigate the impact.

We’ll provide an update in 30 minutes or sooner if additional information becomes available.

Starting from approximately 06:58 UTC on April 17, a subset of customers are experiencing a feature degradation with Agentforce. Customers receive an internal Salesforce error when attempting to perform knowledge-grounded actions using Agentforce via a “Connect API”. Initial investigations have revealed that the trigger of the issue is related to a recent patch. The relevant teams have been engaged to investigate this issue.

We’ll provide an update in 30 minutes or sooner if additional information becomes available.